R Notebook

# Required
install.packages("doParallel")

trying URL 'https://cran.rstudio.com/bin/macosx/el-capitan/contrib/3.6/doParallel_1.0.15.tgz'
Content type 'application/x-gzip' length 196281 bytes (191 KB)
==================================================
downloaded 191 KB

The downloaded binary packages are in
    /var/folders/hm/2md7sccd0479bw81zsh0yyq80000gn/T//RtmptjktGz/downloaded_packages

library(doParallel)

Loading required package: foreach
Loading required package: iterators
Loading required package: parallel

# Find how many cores are on your machine
detectCores() # Result = Typically 4 to 6

[1] 4

# Create Cluster with desired number of cores. Don't use them all! Your computer is running other processes. 
cl <- makeCluster(2)

# Register Cluster
registerDoParallel(cl)

# Confirm how many cores are now "assigned" to R and RStudio
getDoParWorkers() # Result 2 

[1] 2

# Stop Cluster. After performing your tasks, stop your cluster. 
stopCluster(cl)

install.packages("caret")

trying URL 'https://cran.rstudio.com/bin/macosx/el-capitan/contrib/3.6/caret_6.0-85.tgz'
Content type 'application/x-gzip' length 6264988 bytes (6.0 MB)
==================================================
downloaded 6.0 MB

The downloaded binary packages are in
    /var/folders/hm/2md7sccd0479bw81zsh0yyq80000gn/T//RtmptjktGz/downloaded_packages

install.packages("plotly")

trying URL 'https://cran.rstudio.com/bin/macosx/el-capitan/contrib/3.6/plotly_4.9.2.tgz'
Content type 'application/x-gzip' length 2987045 bytes (2.8 MB)
==================================================
downloaded 2.8 MB

The downloaded binary packages are in
    /var/folders/hm/2md7sccd0479bw81zsh0yyq80000gn/T//RtmptjktGz/downloaded_packages

library(caret)

Loading required package: lattice
Loading required package: ggplot2
Registered S3 method overwritten by 'dplyr':
  method           from
  print.rowwise_df     
Registered S3 method overwritten by 'data.table':
  method           from
  print.data.table     

library(plotly)

Registered S3 method overwritten by 'htmlwidgets':
  method           from         
  print.htmlwidget tools:rstudio

Attaching package: ‘plotly’

The following object is masked from ‘package:ggplot2’:

    last_plot

The following object is masked from ‘package:stats’:

    filter

The following object is masked from ‘package:graphics’:

    layout

largeMatrix <- read.csv("LargeMatrix.csv")

# no nas present
sum(is.na(largeMatrix))

[1] 0

iphone_smallMatrix <- read.csv("iphone_smallmatrix_labeled_8d.csv")
samsung_smallMatrix <- read.csv("galaxy_smallmatrix_labeled_9d.csv")

names(iphone_smallMatrix)

 [1] "iphone"          "samsunggalaxy"   "sonyxperia"      "nokialumina"     "htcphone"       
 [6] "ios"             "googleandroid"   "iphonecampos"    "samsungcampos"   "sonycampos"     
[11] "nokiacampos"     "htccampos"       "iphonecamneg"    "samsungcamneg"   "sonycamneg"     
[16] "nokiacamneg"     "htccamneg"       "iphonecamunc"    "samsungcamunc"   "sonycamunc"     
[21] "nokiacamunc"     "htccamunc"       "iphonedispos"    "samsungdispos"   "sonydispos"     
[26] "nokiadispos"     "htcdispos"       "iphonedisneg"    "samsungdisneg"   "sonydisneg"     
[31] "nokiadisneg"     "htcdisneg"       "iphonedisunc"    "samsungdisunc"   "sonydisunc"     
[36] "nokiadisunc"     "htcdisunc"       "iphoneperpos"    "samsungperpos"   "sonyperpos"     
[41] "nokiaperpos"     "htcperpos"       "iphoneperneg"    "samsungperneg"   "sonyperneg"     
[46] "nokiaperneg"     "htcperneg"       "iphoneperunc"    "samsungperunc"   "sonyperunc"     
[51] "nokiaperunc"     "htcperunc"       "iosperpos"       "googleperpos"    "iosperneg"      
[56] "googleperneg"    "iosperunc"       "googleperunc"    "iphonesentiment"

names(samsung_smallMatrix)

 [1] "iphone"          "samsunggalaxy"   "sonyxperia"      "nokialumina"     "htcphone"       
 [6] "ios"             "googleandroid"   "iphonecampos"    "samsungcampos"   "sonycampos"     
[11] "nokiacampos"     "htccampos"       "iphonecamneg"    "samsungcamneg"   "sonycamneg"     
[16] "nokiacamneg"     "htccamneg"       "iphonecamunc"    "samsungcamunc"   "sonycamunc"     
[21] "nokiacamunc"     "htccamunc"       "iphonedispos"    "samsungdispos"   "sonydispos"     
[26] "nokiadispos"     "htcdispos"       "iphonedisneg"    "samsungdisneg"   "sonydisneg"     
[31] "nokiadisneg"     "htcdisneg"       "iphonedisunc"    "samsungdisunc"   "sonydisunc"     
[36] "nokiadisunc"     "htcdisunc"       "iphoneperpos"    "samsungperpos"   "sonyperpos"     
[41] "nokiaperpos"     "htcperpos"       "iphoneperneg"    "samsungperneg"   "sonyperneg"     
[46] "nokiaperneg"     "htcperneg"       "iphoneperunc"    "samsungperunc"   "sonyperunc"     
[51] "nokiaperunc"     "htcperunc"       "iosperpos"       "googleperpos"    "iosperneg"      
[56] "googleperneg"    "iosperunc"       "googleperunc"    "galaxysentiment"

#summary(iphone_smallMatrix$ios)
str(iphone_smallMatrix)

'data.frame':   12973 obs. of  59 variables:
 $ iphone         : int  1 1 1 1 1 41 1 1 1 1 ...
 $ samsunggalaxy  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonyxperia     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokialumina    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcphone       : int  0 0 0 0 0 0 0 0 0 0 ...
 $ ios            : int  0 0 0 0 0 6 0 0 0 0 ...
 $ googleandroid  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonecampos   : int  0 0 0 0 0 1 1 0 0 0 ...
 $ samsungcampos  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonycampos     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiacampos    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htccampos      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonecamneg   : int  0 0 0 0 0 3 1 0 0 0 ...
 $ samsungcamneg  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonycamneg     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiacamneg    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htccamneg      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonecamunc   : int  0 0 0 0 0 7 1 0 0 0 ...
 $ samsungcamunc  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonycamunc     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiacamunc    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htccamunc      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonedispos   : int  0 0 0 0 0 1 13 0 0 0 ...
 $ samsungdispos  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonydispos     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiadispos    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcdispos      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonedisneg   : int  0 0 0 0 0 3 10 0 0 0 ...
 $ samsungdisneg  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonydisneg     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiadisneg    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcdisneg      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonedisunc   : int  0 0 0 0 0 4 9 0 0 0 ...
 $ samsungdisunc  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonydisunc     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiadisunc    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcdisunc      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphoneperpos   : int  0 1 0 1 1 0 5 3 0 0 ...
 $ samsungperpos  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonyperpos     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiaperpos    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcperpos      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphoneperneg   : int  0 0 0 0 0 0 4 1 0 0 ...
 $ samsungperneg  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonyperneg     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiaperneg    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcperneg      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphoneperunc   : int  0 0 0 1 0 0 5 0 0 0 ...
 $ samsungperunc  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonyperunc     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiaperunc    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcperunc      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iosperpos      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ googleperpos   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iosperneg      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ googleperneg   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iosperunc      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ googleperunc   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonesentiment: int  0 0 0 0 0 4 4 0 0 0 ...

str(samsung_smallMatrix)

'data.frame':   12911 obs. of  59 variables:
 $ iphone         : int  1 1 1 0 1 2 1 1 4 1 ...
 $ samsunggalaxy  : int  0 0 1 0 0 0 0 0 0 0 ...
 $ sonyxperia     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokialumina    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcphone       : int  0 0 0 1 0 0 0 0 0 0 ...
 $ ios            : int  0 0 0 0 0 0 0 0 0 0 ...
 $ googleandroid  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonecampos   : int  0 0 1 0 0 1 0 0 0 0 ...
 $ samsungcampos  : int  0 0 1 0 0 0 0 0 0 0 ...
 $ sonycampos     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiacampos    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htccampos      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonecamneg   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ samsungcamneg  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonycamneg     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiacamneg    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htccamneg      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonecamunc   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ samsungcamunc  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonycamunc     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiacamunc    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htccamunc      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonedispos   : int  0 1 0 0 0 0 2 0 0 0 ...
 $ samsungdispos  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonydispos     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiadispos    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcdispos      : int  0 0 0 1 0 0 0 0 0 0 ...
 $ iphonedisneg   : int  0 1 0 0 0 0 0 0 0 0 ...
 $ samsungdisneg  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonydisneg     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiadisneg    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcdisneg      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iphonedisunc   : int  0 1 0 0 0 0 0 0 0 0 ...
 $ samsungdisunc  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonydisunc     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiadisunc    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcdisunc      : int  0 0 0 1 0 0 0 0 0 0 ...
 $ iphoneperpos   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ samsungperpos  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonyperpos     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiaperpos    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcperpos      : int  0 0 0 1 0 0 0 0 0 0 ...
 $ iphoneperneg   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ samsungperneg  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonyperneg     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiaperneg    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcperneg      : int  0 0 0 1 0 0 0 0 0 0 ...
 $ iphoneperunc   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ samsungperunc  : int  0 0 0 0 0 0 0 0 0 0 ...
 $ sonyperunc     : int  0 0 0 0 0 0 0 0 0 0 ...
 $ nokiaperunc    : int  0 0 0 0 0 0 0 0 0 0 ...
 $ htcperunc      : int  0 0 0 1 0 0 0 0 0 0 ...
 $ iosperpos      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ googleperpos   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iosperneg      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ googleperneg   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ iosperunc      : int  0 0 0 0 0 0 0 0 0 0 ...
 $ googleperunc   : int  0 0 0 0 0 0 0 0 0 0 ...
 $ galaxysentiment: int  5 3 3 0 1 0 3 5 5 5 ...

# no missing values
sum(is.na(iphone_smallMatrix))

[1] 0

sum(is.na(samsung_smallMatrix))

[1] 0

# There are no empty columns
colSums(iphone_smallMatrix)

         iphone   samsunggalaxy      sonyxperia     nokialumina        htcphone             ios 
          27867             923             312              30            1779            1976 
  googleandroid    iphonecampos   samsungcampos      sonycampos     nokiacampos       htccampos 
            514            3757             697             129              89            1469 
   iphonecamneg   samsungcamneg      sonycamneg     nokiacamneg       htccamneg    iphonecamunc 
           3044             710              30              80            1206            3289 
  samsungcamunc      sonycamunc     nokiacamunc       htccamunc    iphonedispos   samsungdispos 
            243              42              47             611            9304             819 
     sonydispos     nokiadispos       htcdispos    iphonedisneg   samsungdisneg      sonydisneg 
            148             129            2051            7894             856             161 
    nokiadisneg       htcdisneg    iphonedisunc   samsungdisunc      sonydisunc     nokiadisunc 
            103            1509            6550             352              59              64 
      htcdisunc    iphoneperpos   samsungperpos      sonyperpos     nokiaperpos       htcperpos 
            812            7854             803             112             138            1660 
   iphoneperneg   samsungperneg      sonyperneg     nokiaperneg       htcperneg    iphoneperunc 
           6313             832              83             133            1562            3880 
  samsungperunc      sonyperunc     nokiaperunc       htcperunc       iosperpos    googleperpos 
            287              30              72             832             768             500 
      iosperneg    googleperneg       iosperunc    googleperunc iphonesentiment 
            747             755             429             209           48318 

colSums(samsung_smallMatrix)

         iphone   samsunggalaxy      sonyxperia     nokialumina        htcphone             ios 
          28525             921             311              30            1775            1972 
  googleandroid    iphonecampos   samsungcampos      sonycampos     nokiacampos       htccampos 
            513            3733             693             125              89            1461 
   iphonecamneg   samsungcamneg      sonycamneg     nokiacamneg       htccamneg    iphonecamunc 
           3027             710              30              80            1202            3278 
  samsungcamunc      sonycamunc     nokiacamunc       htccamunc    iphonedispos   samsungdispos 
            241              40              47             608            9251             814 
     sonydispos     nokiadispos       htcdispos    iphonedisneg   samsungdisneg      sonydisneg 
            143             129            2039            7849             856             161 
    nokiadisneg       htcdisneg    iphonedisunc   samsungdisunc      sonydisunc     nokiadisunc 
            103            1501            6519             351              58              64 
      htcdisunc    iphoneperpos   samsungperpos      sonyperpos     nokiaperpos       htcperpos 
            810            7786             801             110             138            1653 
   iphoneperneg   samsungperneg      sonyperneg     nokiaperneg       htcperneg    iphoneperunc 
           6269             832              83             133            1559            3847 
  samsungperunc      sonyperunc     nokiaperunc       htcperunc       iosperpos    googleperpos 
            285              28              72             829             768             500 
      iosperneg    googleperneg       iosperunc    googleperunc galaxysentiment 
            747             755             429             209           49430 

There are multiple phones in this dataset.

plot_ly(iphone_smallMatrix, x= ~iphone_smallMatrix$iphonesentiment, type='histogram')

plot_ly(samsung_smallMatrix, x= ~samsung_smallMatrix$galaxysentiment, type='histogram')

NA

0: very negative

1: negative

2: somewhat negative

3: somewhat positive

4: positive

5: very positive

Interesting.

We have two datasets. iphone and samsung. iphone has all the same variables, except for the variable “iphonesentiment”. the galaxy dataset is similiar with “galaxysentiment”.

We already have too many innecessary predictors, including google, nokia, htc, and sony phones. We are only looking for samsung galaxy and iphone.

sort(names(iphone_smallMatrix))

 [1] "googleandroid"   "googleperneg"    "googleperpos"    "googleperunc"    "htccamneg"      
 [6] "htccampos"       "htccamunc"       "htcdisneg"       "htcdispos"       "htcdisunc"      
[11] "htcperneg"       "htcperpos"       "htcperunc"       "htcphone"        "ios"            
[16] "iosperneg"       "iosperpos"       "iosperunc"       "iphone"          "iphonecamneg"   
[21] "iphonecampos"    "iphonecamunc"    "iphonedisneg"    "iphonedispos"    "iphonedisunc"   
[26] "iphoneperneg"    "iphoneperpos"    "iphoneperunc"    "iphonesentiment" "nokiacamneg"    
[31] "nokiacampos"     "nokiacamunc"     "nokiadisneg"     "nokiadispos"     "nokiadisunc"    
[36] "nokialumina"     "nokiaperneg"     "nokiaperpos"     "nokiaperunc"     "samsungcamneg"  
[41] "samsungcampos"   "samsungcamunc"   "samsungdisneg"   "samsungdispos"   "samsungdisunc"  
[46] "samsunggalaxy"   "samsungperneg"   "samsungperpos"   "samsungperunc"   "sonycamneg"     
[51] "sonycampos"      "sonycamunc"      "sonydisneg"      "sonydispos"      "sonydisunc"     
[56] "sonyperneg"      "sonyperpos"      "sonyperunc"      "sonyxperia"     

toFilter <- grepl('iphone', colnames(iphone_smallMatrix))
 iphones <- iphone_smallMatrix[toFilter]
colnames(iphones)

 [1] "iphone"          "iphonecampos"    "iphonecamneg"    "iphonecamunc"    "iphonedispos"   
 [6] "iphonedisneg"    "iphonedisunc"    "iphoneperpos"    "iphoneperneg"    "iphoneperunc"   
[11] "iphonesentiment"

*is ios attribute considered part of iphone? I say no, because if it was, then it would have had ios infront of it.

# correlation matrix
#install.packages("corrplot")
#library(corrplot)
#options(max.print=1000000)

corr_matrix <- cor(iphones)

corr_plot <- corrplot(as.matrix(corr_matrix))

corr_plot

                      iphone iphonecampos iphonecamneg iphonecamunc iphonedispos iphonedisneg
iphone           1.000000000   0.07815733   0.49052359  0.750403174   0.05262462  0.175572621
iphonecampos     0.078157326   1.00000000   0.54133997  0.473266316   0.27258655  0.148650674
iphonecamneg     0.490523588   0.54133997   1.00000000  0.643460020   0.26198314  0.346878956
iphonecamunc     0.750403174   0.47326632   0.64346002  1.000000000   0.20900762  0.253253711
iphonedispos     0.052624621   0.27258655   0.26198314  0.209007616   1.00000000  0.868765387
iphonedisneg     0.175572621   0.14865067   0.34687896  0.253253711   0.86876539  1.000000000
iphonedisunc     0.250929821   0.18831003   0.29907429  0.361321734   0.88302623  0.879950578
iphoneperpos    -0.009507666   0.34833242   0.25756896  0.190248578   0.65935383  0.530888336
iphoneperneg     0.013863107   0.15191863   0.30887521  0.113175498   0.63776843  0.640995104
iphoneperunc    -0.016037424   0.18725962   0.21757939  0.174433158   0.66523752  0.570044418
iphonesentiment  0.014858654  -0.02973122  -0.08396314  0.001443485   0.01454682  0.003144905
                iphonedisunc iphoneperpos iphoneperneg iphoneperunc iphonesentiment
iphone            0.25092982 -0.009507666  0.013863107  -0.01603742     0.014858654
iphonecampos      0.18831003  0.348332416  0.151918629   0.18725962    -0.029731217
iphonecamneg      0.29907429  0.257568960  0.308875213   0.21757939    -0.083963139
iphonecamunc      0.36132173  0.190248578  0.113175498   0.17443316     0.001443485
iphonedispos      0.88302623  0.659353827  0.637768430   0.66523752     0.014546824
iphonedisneg      0.87995058  0.530888336  0.640995104   0.57004442     0.003144905
iphonedisunc      1.00000000  0.554364879  0.564479458   0.62392944     0.027172723
iphoneperpos      0.55436488  1.000000000  0.794832452   0.79182763     0.029637900
iphoneperneg      0.56447946  0.794832452  1.000000000   0.75948372    -0.004804058
iphoneperunc      0.62392944  0.791827630  0.759483720   1.00000000     0.037199859
iphonesentiment   0.02717272  0.029637900 -0.004804058   0.03719986     1.000000000

# keep columns only for iphone and samsunggalaxy